Periodontitis continues to be a major cause of tooth loss in adults. Porphromonas gingivalis (Pg) is an etiologic agent of periodontitis, however, the mechanisms by which it contributes to the tissue destruction characteristic of this disease remains unknown. The pathogenic mechanisms used by Pg must be understood at the molecular level before effective strategies to interfere with Pg-mediated tissue destruction can be designed. The long-range goal of this research is to design new Pg-specific anti-microbial agents to prevent the development and/or arrest the progression of periodontitis. Pg makes four closely related proteinase/adhesin/hemagglutinins denoted PrtP, Rgp-1, Rgp-2, and HagA that appear to be important virulence factors of this pathogen. Preliminary studies suggest that interference with their functions interferes with the development of periodontitis in the cynomolgus monkey model of ligature-induced periodontitis. We have developed the first system for expression of catalytically active PrtP, Rgp-1, and Rgp-2 in a heterologous prokaryotic host. It will allow us to use molecular biological and genetic methods to sort out structure/function relationships among these proteins at a level not possible in the Pg background. Five Specific Aims are proposed: 1) Determine the proteinase properties and substrate specificities of recombinant (r)PrtP, rRgp-1, rRgp-2, and rHagA, as well as the ability of each to mediate hemagglutination and bind and adhere to extracellular matrix and plasma proteins; 2) Characterize the structural and functional domains of the three recombinant proteinases and recombinant HagA, by (A) determining the identify of the limit peptides comprising the catalytic domains of rPrtP, rRgp-1, and rRgp-2, as well as the identify of the catalytic Cys and His residues of these proteinases, and (B) determining the limit peptides of the hemagglutinin and/or adhesin domains of rPrtP, rRgp-1, and rHagA; 3) Construct and characterize isogenic strains of Pg defective in the expression of prtP, rgp-1, rgp-2, and hagA, individually and in combination; 4) Use the Bacteroides fragilis expression system and the constructs obtained in Aim 2 and Aim 3 to examine the effects of each of these proteins on the extent of processing of the other three; 5) Examine transcription of these genes in wild type Pg, in Pg knockout mutants, and in the Bf expression system to (A) determine the number and sizes of transcripts made from prtP, rgp-1, rgp-2, and hagA, (B) construct transcript-specific probes, and (C) determine whether Bf is a good model in which to study transcription of Pg virulence genes.